[Doc] Convert Sphinx directives ( {class}, {meth}, {attr}, ...) to MkDocs format for better documentation linking (#18663)

Signed-off-by: Zerohertz <ohg3417@gmail.com>
This commit is contained in:
Hyogeun Oh (오효근)
2025-05-27 01:44:20 -07:00
committed by GitHub
parent 6881107948
commit a68e293cb9
37 changed files with 360 additions and 247 deletions
+6 -4
View File
@@ -492,8 +492,9 @@ class MQLLMEngineClient(EngineClient):
from the LLMEngine to the caller.
Args:
prompt: The prompt to the LLM. See {class}`~vllm.inputs.PromptType`
for more details about the format of each input.
prompt: The prompt to the LLM. See
[`PromptType`][vllm.inputs.PromptType] for more details about
the format of each input.
sampling_params: The sampling parameters of the request.
request_id: The unique id of the request.
lora_request: LoRA request to use for generation, if any.
@@ -561,8 +562,9 @@ class MQLLMEngineClient(EngineClient):
from the LLMEngine to the caller.
Args:
prompt: The prompt to the LLM. See {class}`~vllm.inputs.PromptType`
for more details about the format of each input.
prompt: The prompt to the LLM. See
[`PromptType`][vllm.inputs.PromptType] for more details about
the format of each input.
pooling_params: The pooling parameters of the request.
request_id: The unique id of the request.
lora_request: LoRA request to use for generation, if any.
+11 -8
View File
@@ -42,19 +42,22 @@ HEALTHY_RESPONSE = (pickle.dumps(VLLM_RPC_SUCCESS_STR), )
class MQLLMEngine:
"""A multiprocessing wrapper for {class}`LLMEngine`.
"""A multiprocessing wrapper for
[`LLMEngine`][vllm.engine.llm_engine.LLMEngine].
This class is used to wrap the {class}`LLMEngine` class to enable use
This class is used to wrap the
[`LLMEngine`][vllm.engine.llm_engine.LLMEngine] class to enable use
in concurrnet manner. It runs a background loop and uses zeromq to
receive new requests and stream outputs incrementally via ipc.
The {class}`LLMEngine` generate or encode process is kicked off when a new
RPCProcessRequest is received by the input_socket.
The [`LLMEngine`][vllm.engine.llm_engine.LLMEngine] generate or encode
process is kicked off when a new RPCProcessRequest is received by the
input_socket.
The self.engine_loop checks the input_socket for new requests,
adds them to the LLMEngine if there are any, calls the internal
{class}`LLMEngine.step()`, and sends the RequestOutputs back over
the output_socket.
[`LLMEngine.step()`][vllm.engine.llm_engine.LLMEngine.step], and sends
the RequestOutputs back over the output_socket.
If use_async_sockets is set, the logic associated with reading new
requests from the socket and sending data to the socket is passed
@@ -65,8 +68,8 @@ class MQLLMEngine:
ipc_path: Base path for zeromq interprocess messaging
use_async_sockets: Whether to make send/recv async with GPU
log_requests: Whether to log the requests.
*args: Arguments for {class}`LLMEngine`.
**kwargs: Arguments for {class}`LLMEngine`.
*args: Arguments for [`LLMEngine`][vllm.engine.llm_engine.LLMEngine].
**kwargs: Arguments for [`LLMEngine`][vllm.engine.llm_engine.LLMEngine].
"""
def __init__(self,